|

AEO: How to Optimise for AI

After 20 years in SEO, I’ve seen a lot of changes – multiple algorithm updates (panda, penguin, bert, hummingbird), mobile-first indexing, voice search, core web vitals, reviews.
Getting traffic and rankings has always focused on a combination of technical SEO, content SEO and authority SEO. For the most part, that is still true.
Although lately it has been demoralising watching people use machines to write mass content purely to keep the Google machine happy…

Right now, we’re entering the most volatile period I’ve ever seen.  As a search specialist at a digital agency, I’m constantly being asked:
“What should we be telling our clients about AEO? Is SEO still worth doing?”
Every day my inbox is full of articles proclaiming the death of traditional SEO, debating whether old practices still apply.
For small businesses who may have invested years into content and optimisation, it’s understandably concerning, exhausting and overwhelming.

But the fundamentals – clear, useful content; accessible site structure and well-earned authority – still matter. What’s changing is how that content is discovered and surfaced.
I’m an optimist, so I prefer to see the current AI-driven changes as a move that will reward good quality work rather than the mass keyword-focused content of the past.

Just build user-focused, high-quality creative content that genuinely answers questions and provides value.    

That’s always been the goal of good SEO on a good website. Now it just needs packaging in a way that LLMs and answer engines can understand.

So What Is AEO?

AEO stands for Answer Engine Optimisation. It’s the process of making your content easier for AI-powered platforms – like Google’s AI Overviews, ChatGPT, Bing Copilot, and Perplexity – to understand, extract, and present your site as a trusted source when responding to a user’s question.

Unlike traditional SEO, which aims to rank your website for a specific keyword in a list of ten blue links, AEO is about helping your content be the answer. It’s what allows a paragraph from your blog post, a bullet list from your how-to guide, or a definition from your FAQ to appear inside the AI’s response – without the user necessarily clicking through to your site.

Search is evolving into a more conversational, answer-driven experience. These answer engines rely on:

  • Clear structure
  • Accurate, user-focused information
  • Recognisable topical authority

If your content is written and formatted to directly address common user questions – using proper headings, FAQs, summaries, and list formats – it stands a much better chance of being surfaced in these AI-generated responses.

Think of AEO as the next evolution of SEO: moving beyond ranking to being referenced. Luckily, many of the foundational SEO practices already implemented can set you up for success in this new environment.

Personally I’m enjoying AI – it saves me time reading through 10 sites of varying quality to find the real gems. And helps me clarify my thoughts into a simple written response (although I already thought that was one of my super powers).

But it means your content needs to be not only good, but also structured and visible in the ways that AI models understand.

How to optimise pages for AI

Step 1: You need a content strategy

First ask yourself why are you creating this piece of content. Who are you trying to help? What problem are you solving? What is the call-to-action? Is there enough detail to demonstrate your expertise? Is this a pillar page or supporting page? How often can you write about this topic?

The benefits of a content strategy are:

  • It builds topical depth: Search engines and LLMs favour sites that cover a subject comprehensively, not just superficially.
  • It aids consistent publishing: A content calendar helps you stay on track, rather than posting reactively or sporadically.
  • It reinforces expertise: Clustering content around sub-themes builds authority in your niche.
  • It improves discoverability: Interlinking between articles shows search engines and visitors how your topics relate and creates a web of relevance.
  • It sets you up for repurposing: Once you have structured hubs, it’s easier to turn blog content into videos, newsletters, or social posts. Consistency in your niche builds topical authority.

Step 2: Structure your content for AI visibility

A well-structured page is key. Not just for SEO crawlers, but also for LLMs trying to make sense of your site in milliseconds.
Here’s a sample structure for an AI-Friendly blog post:

  • H1: Your main headline (clearly stating the topic)
  • Opening summary paragraph: What this post is about and why it matters
    • H2: What is [your topic]?
    • H2: Key benefits or advantages
    • H2: Step-by-step guide or how-to
      • H3: Individual step details
      • H3: Next step details
    • H2: Deep dive or real-world application
    • H2: Common questions (FAQ style)
    • H2: Summary takeaways (dot points)

This format helps LLMs extract structured, clean information for users. Even if a user never clicks through, AI may quote your content—which builds authority and trust.

Step 3: Use Schema Markup (Structured Data)

Go beyond the basics. You’ll need your developer involved here (or a good plugin). Use schema markup to help AI engines understand your different types of content:

  • FAQ schema for Q&As
  • How To tags for step-by-step guides
  • Article/BlogPosting schema for regular posts
  • Product/Merchant schema for products to appear in shopping engines
  • LocalBusiness schema for individual store locations
  • Organisation schema for businesses
  • Video schema for video search

Step 4: Focus on user experience

Congratulations. You appeared in an AI answer and they clicked through to your site.
Now make the website experience SO good, it makes them want to come back, read more articles, signup for your newsletter, buy your product.

  • Refresh your theme – make your site looked modern and cared-for.
  • Provide a summary at the top – they may be short of time, and you only have a few seconds to get their attention/interest anyway.
  • Demonstrate your topic expertise. Show your visitor they’ve come to the right place.
  • Create clear, concise content that’s easy to read
  • Break up large blocks of content with images, videos, infographics.
  • Publish regularly – at least one blog post per month, and preferably more, to show that you stay up-to-date.

Step 5: Traditional SEO practices still matter

Don’t need to throw out your SEO playbook. But you do need to adjust it for the way AI is now evaluating content. These traditional factors do still matter:

  • Domain Authority: Try to get links from trusted domains to feed AI confidence. Make sure you exist on key industry, business and local directories
  • Internal Linking: Build topical relationships and clarity
  • Keyword Relevance: Focus on topic authority rather than keyword density
  • Backlinks: Only helpful if high quality and relevant
  • Content Freshness: Timely updates keep content visible
  • Page Load & Accessibility: Helps humans and crawlers alike

Add extra content types such as images, videos and PDFs so that you can be found via image search, and video search.
Prepare your images wisely, with meaningful filenames, small filesizes to load quickly, concise image alt tags for discoverability and accessibility.
Prepare your videos wisely, with meaningful video names, descriptions, thumbnails and transcripts.
PDFs do get indexed, so use meaningful filenames and titles, and provide meta tags inside the PDF.

Local SEO is still huge. Claim your local business profile for every location, on Google and Business.
Maximise the content, filling in every possible field, service type and service area.
Upload photos of the outside and inside of your business.
Respond to reviews.
Use updates, promotions and product sections of the local listing.

Step 6: Get people talking and sharing

AI doesn’t just rely on traditional links from authority sites.
Your brand might be gaining visibility even without actual clickable backlinks – as long as it’s being mentioned in the right places – on public, crawlable platforms. Even better when you have a recognisable brand name that doesn’t rely on links to be found.
You don’t have to go viral – although it definitely helps. Historically good SEOs have always recommended creating such awesome content that people voluntarily share it as a quality resource, on:

  • Public Facebook and Instagram Pages
  • LinkedIn
  • Reddit
  • Quora
  • X (Twitter)
  • Medium
  • Substack
  • Industry blog posts
  • Wikipedia

It can be hard when you start out and nobody knows you exist yet.
Consider paid advertising to raise brand awareness and generate those lovely mentions and links on other sites:

  • Paid Google Ads (Display and Video Ads for maximum affordable awareness and reach)
  • Paid Meta Ads (Use multiple formats (display, carousel, video for relevant audiences)
  • LinkedIn Ads (For B2B businesses)

Reviews are more important than ever.
Make sure you claim your website profile on Trustpilot and other review platforms.
Respond to every negative review and offer to resolve their issues. Thank people for positive reviews.

Step 7: Measure and monitor your successes

  • Start with a Measurement Plan – documenting your objectives, KPIs, analytics approach, conversion goals, campaign naming standards.
  • Implement Analytics tracking – whether it’s Google Analytics, Matomo or other – track your traffic sources, page engagement and conversions
  • Build a reporting dashboard – it’s the most efficient way to summarise how your site is performing (traffic, conversions, engagement)
  • Focus on conversions rather than keyword rankings. That’s the key thing changing for SEOs. Your Google rankings WILL be dropping as people move to AI.
  • Track brand mentions and sentiment. Perform a brand audit – does AI see you the way you intended? Are people saying positive or negative things about you?
  • Use a heatmap. When people come to a page, what do they click on? What do they scroll past? Are there any rage clicks?
  • Track which pages receive traffic from AI. Analyse what worked, can it be improved further and applied to other pages.
  • Monitor page engagement. If people bounce (leave the site after visiting one page), can that page be improved? Or is it not worth saving, and should be redirected to a better page?

Summary

AI in search rewards clarity, structure, and expertise. Focus on human-first content that’s easy for people to read, and for machines to understand.
If you build it (properly), they will come :)

For fun, here’s a terrible diagram I butchered in Canva. It can only get better from here.
x